WO2023217032A1 - 一种机器人碰撞检测方法、存储介质及电子设备 - Google Patents

一种机器人碰撞检测方法、存储介质及电子设备 Download PDF

Info

Publication number
WO2023217032A1
WO2023217032A1 PCT/CN2023/092493 CN2023092493W WO2023217032A1 WO 2023217032 A1 WO2023217032 A1 WO 2023217032A1 CN 2023092493 W CN2023092493 W CN 2023092493W WO 2023217032 A1 WO2023217032 A1 WO 2023217032A1
Authority
WO
WIPO (PCT)
Prior art keywords
robot
collision detection
universal cylinder
target
cylinder
Prior art date
Application number
PCT/CN2023/092493
Other languages
English (en)
French (fr)
Inventor
谢胜文
王珂
Original Assignee
苏州艾利特机器人有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 苏州艾利特机器人有限公司 filed Critical 苏州艾利特机器人有限公司
Publication of WO2023217032A1 publication Critical patent/WO2023217032A1/zh

Links

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00—Programme-controlled manipulators
    • B25J9/16—Programme controls
    • B25J9/1602—Programme controls characterised by the control system, structure, architecture
    • B25J9/161—Hardware, e.g. neural networks, fuzzy logic, interfaces, processor
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00—Programme-controlled manipulators
    • B25J9/16—Programme controls
    • B25J9/1612—Programme controls characterised by the hand, wrist, grip control
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00—Programme-controlled manipulators
    • B25J9/16—Programme controls
    • B25J9/1656—Programme controls characterised by programming, planning systems for manipulators
    • B25J9/1664—Programme controls characterised by programming, planning systems for manipulators characterised by motion, path, trajectory planning
    • B25J9/1666—Avoiding collision or forbidden zones

Definitions

  • the invention relates to the field of robot technology, and in particular to a collision detection method, a storage medium and an electronic device for a robot.
  • the robot can automatically perform work according to the preset program. During the operation of the robot, there is less user intervention, and the safety of its operation needs to be ensured. For the safety of the robot, the robot needs to be able to perform collision detection timely and accurately, including the collision of the robot itself, the collision of the robot body and the load connected to the robot, and the possible collision of the robot and objects in the environment.
  • the detection of robot collisions in the existing technology is mostly based on sensors or motors. For example, when a collision occurs, the joint current of the robot increases to determine that a collision may occur. However, this method has limited accuracy and is prone to misjudgment; or, The electronic skin is installed on the robot, and the electronic skin fully covers the robot body, so that collisions occurring in any part of the robot can be identified. However, the installation cost of the electronic skin is high, the wiring is complex, and the anti-interference effect is poor.
  • the purpose of the present invention is to provide a robot collision detection method, storage medium and electronic equipment to solve the problems of low robot collision detection accuracy and difficult configuration in the existing technology.
  • the robot collision detection method provided by the invention can improve the shape of the robot.
  • the fitting accuracy is high, the collision point and puncture depth can be known at the same time, and the calculation efficiency is high.
  • a robot collision detection method including:
  • the universal cylinder includes an upper bottom surface, a lower bottom surface and a cylinder surface.
  • the upper bottom surface is parallel to the lower bottom surface.
  • the line connecting the center points of the upper bottom surface and the lower bottom surface is perpendicular to the upper bottom surface and the lower bottom surface.
  • the general cylinder parameters include the major axis radius a 1 and minor axis radius b 1 of the upper bottom surface, the major axis radius a 2 and minor axis radius b 2 of the lower bottom surface, and the cylinder height h;
  • the GJK algorithm is used for collision detection of the robot.
  • a second aspect of the present application provides a computer-readable storage medium that stores a computer program.
  • the computer program When executed by a processor, it implements any of the robot collision detection methods described above.
  • a third aspect of the present application provides an electronic device, including: a memory storing a computer program; a processor configured to execute the computer program in the memory to implement any one of the above Robot collision detection method.
  • the beneficial effects of the specific embodiments of the present invention are at least as follows: fitting the robot through a universal cylinder, the fitting accuracy of the robot is higher compared with the standard geometry, which is beneficial to improving the accuracy of collision detection; for universal
  • the cylindrical design supports the mapping function and uses the GJK algorithm for collision detection, which has high computational efficiency.
  • the solution provided by this solution is suitable for the collision between a general cylinder and any standard geometry, and has wide applicability.
  • Figure 1 is a schematic diagram of a robot collision detection method according to an embodiment of the present invention
  • Figure 2a is a schematic diagram of a universal cylinder according to an embodiment of the present invention.
  • Figure 2b is a schematic diagram of a universal cylinder according to another embodiment of the present invention.
  • Figure 3 is a schematic diagram of a robot according to an embodiment of the present invention.
  • Figure 4 is a schematic diagram of a universal cylinder fitting robot according to an embodiment of the present invention.
  • FIG. 5 is a block diagram of an electronic device according to an embodiment of the present invention.
  • the present invention protects a collision detection method for a robot.
  • the collision detection method includes:
  • the universal cylinder includes an upper bottom surface, a lower bottom surface and a cylinder surface.
  • the upper bottom surface is parallel to the lower bottom surface.
  • the line connecting the center points of the upper bottom surface and the lower bottom surface is perpendicular to the upper bottom surface and the lower bottom surface.
  • the general cylinder parameters include the major axis radius a1 and minor axis radius b1 of the upper bottom surface, the major axis radius a2 and minor axis radius b2 of the lower bottom surface, and the cylinder height h;
  • Figure 2a is a schematic diagram of a universal cylinder according to an embodiment of the present application.
  • the universal cylinder includes an upper bottom surface, a lower bottom surface and a cylinder surface connecting the upper bottom surface and the lower bottom surface.
  • the upper bottom surface and the lower bottom surface respectively include a long axis and a
  • the minor axis, major axis and minor axis can be of different lengths. Specifically, the major axis radius of the upper bottom surface a 1 ⁇ the minor axis radius of the upper bottom surface b 1 , the major axis radius of the lower base a 2 ⁇ the minor axis radius of the lower base b 2 , so that the upper bottom surface and the lower bottom surface are respectively formed into ellipses.
  • the universal cylinder can be adjusted by modifying the parameters of the universal cylinder to achieve shape fitting of the robot.
  • the upper bottom surface is parallel to the lower bottom surface, and the line connecting the center points of the upper bottom surface and the lower bottom surface is perpendicular to the upper bottom surface and the lower bottom surface, so that the universal cylinder appears as a convex cylinder and avoids collision detection through the GJK algorithm. An exception occurs.
  • Figure 2b exemplarily shows a schematic diagram of another universal cylinder.
  • the universal cylinder is easier to accurately achieve shape fitting when dealing with scenes with irregular circular cross-sections and differences in parameters between the upper and lower bottom surfaces.
  • the robot is configured to perform collision detection, the robot is configured to include a geometric shape fitting library, and selects a geometric shape to fit the part to be fitted according to the geometric shape fitting library, wherein the geometric shape fitting library includes
  • the universal cylinder and standard geometry include at least some of spheres, cylinders, cubes and cones.
  • Figure 3 shows an example of a robot body configuration.
  • the robot body 100 includes a base 10, joints 20 and connecting rods 30.
  • the main components of the robot can be fitted through standard cylinders, but when the robot When the component does not present a standard cylindrical shape, the cylinder fitting accuracy is poor, which will affect the collision detection accuracy of the robot; when a universal cylinder is used, compared with traditional cylinders and other shapes, the universal cylinder includes more parameter variables.
  • the universal cylinder can be presented as a cylindrical configuration with an elliptical upper and lower base, or a cylindrical configuration with different sizes of the upper and lower bases, see Figure 4 schematically shows a schematic diagram of fitting a robot through a universal cylinder, in which the joints 20 of the robot present an elliptical bottom surface.
  • the upper and lower surfaces can be The bottom surface and the lower bottom surface present different shapes, so that the robot joints can be fitted.
  • the robot connecting rod 30 can also be fitted.
  • universal cylinder fitting it can be closer to the shape of the robot itself and improve the accuracy of robot shape fitting.
  • two universal cylinders are used to realize the robot connection. rod fitting.
  • step S2 determines the target fitting part of the universal cylinder including: S21. Obtaining a structural model of the robot and/or the robot working environment; S22. Splitting the structural model into multiple parts to be fitted; S23. According to the required The part to be fitted is determined to be a target fitting part suitable for shape fitting based on the general cylinder.
  • the options include the robot body, the robot load, and at least part of the robot working environment.
  • Structural model split the structural model, for example, according to different component parts, or according to parts of different shapes, so as to select from the parts to be fitted suitable for fitting by a universal cylinder.
  • the target fitting part of the combination is
  • the robot includes a robot body and, optionally, a load connected to the robot body.
  • Obtaining the structural model of the robot includes acquiring the body structure model of the robot, and/or acquiring the load structure model connected to the robot body. It should be noted that in this article, when the concept of "robot” is mentioned, it includes the robot body and the entire robot after the robot body is connected to the load. The specific meaning should be understood as a whole according to the context.
  • step S23 determines the target fitting part suitable for shape fitting based on the universal cylinder according to the part to be fitted, including: obtaining the user's operation instruction to determine the target fitting part, and the operation instruction is based on the selection of the part to be fitted.
  • the target fitting part for example, the robot also includes a teaching pendant, through which the robot can be manipulated and controlled.
  • the teaching pendant includes a user interactive interface to present the structural model of the robot, and the user can input operating instructions to control the robot based on the teaching pendant.
  • the robot is configured as a plug-in with shape selection, through which the target fitting part can be automatically determined according to the shape selection method configured by the plug-in.
  • the preset shape selection method exemplarily includes: determining the shape matching degree of the part to be fitted and the universal cylinder, and determining the part to be fitted that meets the preset matching conditions as the target object according to the shape matching degree. joint part.
  • preset matching conditions are set in advance, such as setting a matching degree threshold between the part to be fitted and a universal cylinder, and when the threshold is met, the current part to be fitted is fitted according to the universal cylinder; or, the current part to be fitted is fitted.
  • the matching degree of the part to be fitted is calculated with all the geometries in the geometric shape fitting library. If the matching degree between the current part to be fitted and the universal cylinder is the best, then the shape fitting of the current part to be fitted is performed based on the universal cylinder. That is, the default shape selection at this time
  • the method is to determine whether to fit the current part to be fitted based on the universal cylinder based on whether the matching degree between the current part to be fitted and the universal cylinder is optimal.
  • the universal cylinder parameters can be adjusted through the control page of the user and/or the robot manufacturer so that the matching degree between the universal cylinder and the target fitting part is as high as possible.
  • the universal cylinder When the universal cylinder When the cylinder can fit the target fitting part well, that is, the universal cylinder can surround the target fitting part, and when the volume of the universal cylinder is as small as possible, the parameters of the universal cylinder that have currently completed the fitting are determined as the target parameters, that is, based on the target The parameters can achieve shape fitting of the target fitting part and ensure the accuracy of the fitting process.
  • step S2 determines the target fitting part of the universal cylinder also includes: when fitting the body of the robot, obtaining the zero posture of the robot, and determining the target fitting part according to the robot in the zero position.
  • Libccd is an open source algorithm for detecting collisions. It can perform efficient collision detection on convex geometric shapes. When performing collision detection based on this method, it is necessary to design a supporting mapping function to describe the shape of the geometry. In the existing technology, any complex shape can be described through point clouds, and the computational complexity of calculating its supporting mapping function is linear. However, when using point clouds to describe geometric shapes, if you want to achieve real-time collision detection, it consumes There are many computing resources, and the computing efficiency will be relatively low. By designing corresponding support mapping functions based on reasonable descriptions of geometric shapes, the problems of detection accuracy and calculation efficiency can be effectively solved.
  • the support mapping functions For example, for some basic geometric shapes, such as spheres, standard cylinders, cubes, etc., the support mapping functions The calculation is relatively simple and consumes less computing resources. However, when the shape of the target fitting part that needs to be fitted is not a standard geometry, the accuracy of fitting through the standard geometry is relatively poor, which will lead to an increase in collision detection errors. big. In this application, the concept of a universal cylinder is proposed. Compared with traditional cylinders, it has more parameters for adjustment and can better adapt to robots of different shapes. Through reasonable design of corresponding support mapping functions, detection accuracy can be taken into account and computational efficiency.
  • step S3 calculates the support mapping function of the universal cylinder according to the target parameters including:
  • the general cylinder in Figure 2a exemplifies the construction of a local coordinate system.
  • the conversion direction w of the local coordinate system, the preset conversion direction that is, the conversion relationship between the robot base coordinate system and the local coordinate system of the universal cylinder, can be preset by the robot manufacturer to convert the target direction in the robot base coordinate system Go to the local coordinate system of the general cylinder, find the farthest projected point M, and then convert the farthest projected point into the target coordinates based on the robot's base coordinate system.
  • the support point S C (v) corresponding to the N prism must be At the vertex of the prism, similarly, for a universal cylinder, its support points must exist on the upper base and/or lower surface.
  • step S32 determines the farthest point M of the projection of the transformation direction w based on the local coordinate system, including:
  • Step S4 is executed.
  • the GJK algorithm is used to perform collision detection of the robot.
  • the GJK (Gilbert-johnson-Keerthi) algorithm is a collision detection algorithm that can realize collision detection between convex geometric shapes in space.
  • the GJK algorithm relies on the construction of a support mapping function and determines the relationship between two polygons/volumes. Whether there are two coordinates that are subtracted as the origin is used to determine whether a collision occurs.
  • GJK relies on the process of supporting mapping function calculations to have high computational efficiency and consume little resources.
  • step S4 uses the GJK algorithm to perform collision detection on the robot, it also includes: S41. Determine other parts to be fitted except the target fitting part in the parts to be fitted of the robot, and perform other parts to be fitted according to the standard geometry. Shape fitting, determine the shape fitting parameters of other parts to be fitted and calculate the corresponding support mapping function. After the robot completes shape fitting of all parts to be fitted, it can perform collision detection based on the corresponding support mapping function and GJK algorithm. Among them, the calculation of the support mapping function of standard geometry has been involved in the existing technology, and will not be expanded here. All the After the fitting process is completed, a set of fitting shapes for collision detection is generated, thereby performing collision detection according to the GJK algorithm.
  • step S4 uses the GJK algorithm to perform collision detection on the robot, it also includes: generating a set of fitting shapes for robot collision detection based on the shape fitting of the parts to be fitted, determining two of the shapes as collision detection objects, and using the GJK algorithm to In the three-dimensional space, it is detected whether the two shapes overlap, intersect, or separate. When overlapping or intersecting is detected, it is determined that a collision has occurred.
  • the shape fitting of the robot includes the optional fitting of the robot body, the load connected to the body, and the robot environment. Shape fitting is generated based on all the fitting results, and each time the shape fitting is performed in the fitted shape set. Perform collision detection on two shapes to determine whether there is a collision between the two shapes.
  • the above preferred embodiment of this solution defines the type of universal cylinder.
  • the shape of the upper and lower bottom surfaces of the universal cylinder is easy to adjust and is more adaptable to robot structures on the market, especially for collaborative robots.
  • the structure and accurate fitting of the robot can ensure the accuracy of collision detection.
  • a support mapping function for general cylinders is designed, which has high computational efficiency and consumes little resources.
  • the collision detection method provided by this embodiment does not rely on the detection results of the sensor, and simplifies the robot hardware requirements through accurate shape fitting, while ensuring the accuracy of collision detection.
  • the present application also provides a computer-readable storage medium in which a computer program is stored, such as a memory in which a computer program is stored.
  • the computer program can be executed by a processor to complete a robot collision detection method.
  • the storage medium may be a non-transitory computer-readable storage medium, for example, the non-transitory computer-readable storage medium may be ROM, random access memory (RAM), CD-ROM, magnetic tape, floppy disk, and optical data storage Equipment etc.
  • the present application also provides an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program; the processor is used to execute the computer program in the memory, To implement each step of the above-mentioned robot collision detection method.
  • electronic device 500 may include a processor 510, a memory 520, an input/output component 530, and a communication port 540.
  • Processor (eg, CPU) 510 may execute program instructions in the form of one or more processors.
  • Memory 520 includes different forms of program memory and data storage, such as hard disk, read-only memory (ROM), random access memory (RAM), etc., used to store various data files processed and/or transmitted by the computer.
  • Input/output component 530 may be used to support input/output between the processing device and other components.
  • the communication port 540 can be connected to the network for data communication.
  • Exemplary processing devices may include program instructions stored in read-only memory (ROM), random access memory (RAM), and/or other types of non-transitory storage media for execution by processor 510 .
  • ROM read-only memory
  • RAM random access memory
  • exemplary processing devices may include program instructions stored in read-only memory (ROM), random access memory (RAM), and/or other types of non-transitory storage media for execution by processor 510 .
  • ROM read-only memory
  • RAM random access memory
  • non-transitory storage media for execution by processor 510 .
  • the methods and/or processes of the embodiments of this specification can be implemented in the form of program instructions.

Landscapes

  • Engineering & Computer Science (AREA)
  • Robotics (AREA)
  • Mechanical Engineering (AREA)
  • Automation & Control Theory (AREA)
  • Physics & Mathematics (AREA)
  • Orthopedic Medicine & Surgery (AREA)
  • Health & Medical Sciences (AREA)
  • Artificial Intelligence (AREA)
  • General Health & Medical Sciences (AREA)
  • Evolutionary Computation (AREA)
  • Fuzzy Systems (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Manipulator (AREA)

Abstract

一种机器人碰撞检测方法，该方法包括：S1、预先定义通用圆柱和通用圆柱参数，通用圆柱包括上底面、下底面和柱面，上底面平行于下底面，上底面和下底面的中心点连线垂直于上底面和下底面，通用圆柱包括上底面的长轴半径a1和短轴半径b1、下底面的长轴半径a2和短轴半径b2、以及柱面高度h；S2、确定通用圆柱的目标拟合部位，调整通用圆柱参数以对目标拟合部位的形状进行拟合，将完成拟合时的通用圆柱参数确定为目标参数；S3、根据目标参数计算通用圆柱的支撑映射函数；S4、根据通用圆柱的支撑映射函数，采用GJK算法进行机器人的碰撞检测。该方法简化机器人硬件配置，提升了碰撞检测的精度且计算效率高。还提供一种计算机可读存储介质及电子设备。

Description

一种机器人碰撞检测方法、存储介质及电子设备 技术领域
本发明涉及机器人技术领域，特别是涉及一种机器人的碰撞检测方法、存储介质及电子设备。
背景技术
机器人能够根据预设的程序自动的执行工作，在机器人运行过程中，较少出现用户的干预，需要保证其运行的安全性。出于对机器人安全性的考量，需要机器人能够及时、准确的进行碰撞检测，包括机器人本体自身发生碰撞、机器人本体与机器人所连接的负载的碰撞以及机器人与环境中物体可能的碰撞。
现有技术中对机器人碰撞的检测多基于传感器或电机，例如，通过发生碰撞时，机器人的关节电流增大，以判断可能发生碰撞，但该种方式准确性有限，容易发生误判；或者，通过机器人设置电子皮肤，电子皮肤对机器人本体进行全面覆盖，从而可以识别机器人任何部位发生的碰撞，但是电子皮肤的设置成本较高、布线复杂、且抗干扰效果能力差。
在对本申请技术方案进行研究开发的过程中，发明人发现现有技术中还存在将机器人本体和负载分别拟合为圆柱，通过圆柱与圆柱之间是否存在相交来进行碰撞检测的方式，但是，该种方式只能适用于负载是圆柱的情形，适用性较为有限；同时，机器人本体的形状并非标准圆柱，对机器人本体拟合的精度较低。同时，该方案中只能判断两个圆柱之间是否存在相交的情况，不能判断两者的碰撞点和穿刺深度，不利于机器人后续动作的控制。
发明内容
本发明的目的在于提供一种机器人的碰撞检测方法、存储介质及电子设备，以解决现有技术中机器人碰撞检测精度低、配置难度大的问题，本发明所提供的机器人碰撞检测方法能够提升形状拟合的精度、同时能够获知碰撞点和穿刺深度，且计算效率较高。
为实现上述目标，本申请的技术方案如下：
根据本申请的第一方面，提供一种机器人碰撞检测方法，包括：
S1、预先定义通用圆柱和通用圆柱参数，所述通用圆柱包括上底面、下底面和柱面，所述上底面平行于下底面，上底面和下底面的中心点连线垂直于上底面和下底面，所述通用圆柱参数包括上底面的长轴半径a1和短轴半径b1、下底面的长轴半径a2和短轴半径b2、以及柱面高度h；
S2、确定通用圆柱的目标拟合部位，调整通用圆柱参数以对目标拟合部位的形状进行 拟合，将完成拟合时的通用圆柱参数确定为目标参数；
S3、根据所述目标参数计算通用圆柱的支撑映射函数；
S4、根据通用圆柱的支撑映射函数，采用GJK算法进行机器人的碰撞检测。
本申请的第二方面提供了一种计算机可读存储介质，存储有计算机程序，所述计算机程序被处理器执行时实现前文中任意一项所述的机器人碰撞检测方法。
本申请的第三方面提供了一种电子设备，包括：存储器，所述存储器存储有计算机程序；处理器，用于执行所述存储器中的所述计算机程序，以实现前文中任一项所述机器人碰撞检测方法。
与现有技术相比，本发明具体实施方式的有益效果至少在于：通过通用圆柱对机器人进行拟合，相对于标准几何体对机器人的拟合精度更高，有利于提升碰撞检测的精度；针对通用圆柱设计支撑映射函数，采用GJK算法进行碰撞检测，计算效率高。同时，本方案所提供的方案适用于通用圆柱与任意一种标准几何体之间的碰撞，适用性较广。
附图说明
图1是本发明一个实施例的机器人碰撞检测方法的示意图；
图2a是本发明一个实施例的通用圆柱的示意图；
图2b是本发明另一实施例的通用圆柱的示意图；
图3是本发明一个实施例的机器人的示意图；
图4是本发明一个实施例通过通用圆柱拟合机器人的示意图；
图5是本发明一个实施例的电子设备的框图。
具体实施方式
为使本发明的技术方案更加清楚明了，下面将结合附图来描述本发明的实施例。应当理解的是，对实施方式的具体说明仅用于示教本领域技术人员如何实施本发明，而不是用于穷举本发明的所有可行方式，更不是用于限制本发明的具体实施范围。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动的前提下所获得的所有其他实施例，都应属于本发明保护的范围。
需要说明的是，术语“中心”、“上”、“下”、“前”、“后”、“左”、“右”、“水平”、“顶”、“底”、“竖直”、“水平”、“竖向”等指示的方位或位置关系为基于附图所示的方位或位置关系，仅是为了便于描述或简化描述本发明，而不是指示或暗示所指的装置或元件必须具有特定的方位、以特定的方位构造、安装及操作，因此不能理解为对本发明的限制。此外，在本发明的描述中，“多个”的含义是两个或两个以上，除非另有明确具体的限定。
本发明保护一种机器人的碰撞检测方法，参图1，所述碰撞检测方法包括：
S1、预先定义通用圆柱和通用圆柱参数，所述通用圆柱包括上底面、下底面和柱面，所述上底面平行于下底面，上底面和下底面的中心点连线垂直于上底面和下底面，所述通用圆柱参数包括上底面的长轴半径a1和短轴半径b1、下底面的长轴半径a2和短轴半径b2、以及柱面高度h；
具体的，参图2a，图2a是本申请一个实施例的通用圆柱的示意图，通用圆柱包括上底面、下底面和连接上底面和下底面的柱面，上底面和下底面分别包括长轴和短轴，长轴和短轴可以是不同的长度，具体的，上底面的长轴半径a1≥上底面短轴半径b1,下底面的长轴半径a2≥下底面短轴半径b2，从而使得上底面和下底面分别形成为椭圆，上底面和下底面的参数的改变，可以使得通用圆柱呈现出不同的构型，柱面高度h的变化，可以使得通用圆柱呈现出不同的大小。通用圆柱可以通过修改通用圆柱参数来调整其形状，以实现对机器人的形状拟合。其中，所述上底面平行于下底面，且上底面和下底面的中心点的连线垂直于上底面和下底面，以使得所述通用圆柱呈现为凸面的圆柱，避免通过GJK算法进行碰撞检测时发生异常。图2b示例性的给出了另一种通用圆柱的示意图，通用圆柱在处理截面为不规则圆形，以及处理上底面和下底面参数存在差异的场景时，更容易精确的实现形状拟合。
进一步的，机器人被配置为可进行碰撞检测，机器人被配置为包括几何形状拟合库，根据几何形状拟合库选择几何形状对待拟合部位进行拟合，其中，所述几何形状拟合库包括所述通用圆柱和标准几何体，所述标准几何体包括球体、圆柱体、立方体和锥体中的至少部分。
参图3，图3给出了一个机器人本体构型的示例，机器人本体100包括底座10、关节20和连杆30，通常，可通过标准圆柱对机器人的主要部件进行拟合，但当机器人的部件并非呈现出标准圆柱形状时，通过圆柱拟合精度较差，将会影响机器人的碰撞检测精度；当采用通用圆柱时，相较于传统的圆柱等形状，通用圆柱包括更多的参数变量，通过调整上底面和下底面各自的长轴和短轴长度，通用圆柱可以呈现为上底面和下底面为椭圆的圆柱构型，或者可以呈现为上底面和下底面大小不同的圆柱构型，参图4，图4示例性的展示了通过通用圆柱拟合机器人的示意图，其中机器人的关节20呈现出椭圆的底面，通过调整通用圆柱上底面和下底面的长轴和短轴长度，可以使得上底面和下底面呈现出不同的形状，从而可以实现对机器人关节的拟合，同理，也可以实现对机器人连杆30的拟合。通过通用圆柱拟合的方式，能够更接近机器人本身的形状，提升对机器人形状拟合的精确度，例如图4中，对于不规则的形状的机器人连杆，通过两个通用圆柱实现对机器人连杆 的拟合。
S2、确定通用圆柱的目标拟合部位，调整通用圆柱参数以对目标拟合部位的形状进行拟合，将完成拟合时的通用圆柱参数确定为目标参数；
当对机器人进行碰撞检测时，机器人可能发生碰撞的场景包括：机器人本体自身发生碰撞、机器人本体与机器人所连接的负载间发生碰撞、以及机器人或负载与机器人所处环境发生的碰撞。进一步的，步骤S2确定通用圆柱的目标拟合部位包括：S21、获取机器人和/或机器人工作环境的结构模型；S22、将所述结构模型拆分为多个待拟合部位；S23、根据所述待拟合部位确定适合根据通用圆柱进行形状拟合的目标拟合部位。首先，确定对机器人进行碰撞检测需要关注的对象，根据机器人的工作环境，可选择的包括机器人本体、机器人负载和机器人工作环境中的至少部分，选择性的获取机器人本体、机器人负载以及机器人工作环境的结构模型，将结构模型进行拆分，例如，根据不同的零部件部位进行拆分，或者根据不同的形状的部位进行拆分，从而从待拟合的部位中选择适于通过通用圆柱进行拟合的目标拟合部位。
机器人包括机器人本体，以及可选择的，包括机器人本体所连接的负载，获取机器人的结构模型包括获取机器人的本体结构模型，和/或获取机器人本体所连接的负载结构模型。需要说明的是，本文中，提及“机器人”概念时包括机器人本体，以及机器人本体连接负载后的机器人整体，具体应当根据上下文意思做整体理解。
具体的，步骤S23根据所述待拟合部位确定适合根据通用圆柱进行形状拟合的目标拟合部位包括：获取用户的操作指令以确定目标拟合部位，所述操作指令基于待拟合部位选择目标拟合部位，示例性的，机器人还包括示教器，通过示教器可对机器人进行操纵和控制，示教器包括用户交互界面呈现机器人的结构模型，用户可以通过输入操作指令，从而基于待拟合部位选择目标拟合部位；和/或，根据预设的形状选择方法确定目标拟合部位，所述预设的形状选择方法被配置为基于待拟合部位选择目标拟合部位，示例性的，机器人被配置为具有形状选择的插件，通过该插件可以自动根据该插件配置的形状选择方法确定目标拟合部位。进一步的，所述预设的形状选择方法示例性的包括：确定待拟合部位与通用圆柱的形状匹配度，根据所述形状匹配度将符合预设匹配条件的待拟合部位确定为目标拟合部位。示例性的，预先设定预设匹配条件，例如设定待拟合部位与通用圆柱的匹配度阈值，满足该阈值时根据通用圆柱对当前的待拟合部位进行拟合；或者，对当前的待拟合部位与几何形状拟合库的所有几何体进行匹配度计算，若当前的待拟合部位与通用圆柱的匹配度最好时根据通用圆柱对当前的待拟合部位进行形状拟合，也即此时预设的形状选择 方法为根据当前的待拟合部位与通用圆柱的匹配度是否最佳来确定是否基于通用圆柱拟合当前的待拟合部位。
示例性的，对目标拟合部位拟合的过程中，可以通过用户和/或机器人厂商的控制页面，调整通用圆柱参数以使得通用圆柱与目标拟合部位的匹配度尽可能的高，当通用圆柱能够较好的拟合目标拟合部位时，即通用圆柱能够包围目标拟合部位，并且通用圆柱体积尽可能较小时，将当前完成拟合的通用圆柱参数确定为目标参数，即基于该目标参数能够实现对目标拟合部位的形状拟合，且保证拟合过程的精度。
进一步的，步骤S2确定通用圆柱的目标拟合部位还包括：当对机器人的本体进行拟合时，获取机器人的零位姿态，根据零位状态下的机器人确定目标拟合部位。
S3、根据所述目标参数计算通用圆柱的支撑映射函数；
Libccd是检测碰撞的开源算法，能够对凸的几何形状进行高效的碰撞检测，基于该方法进行碰撞检测时需要针对性设计支撑映射函数以描述几何体的形状。现有技术中，可以通过点云的方式描述任意复杂的形状，而且计算其支撑映射函数的计算复杂度是线性的，但是使用点云描述几何形状时，若想要实现实时的碰撞检测，消耗的计算资源很多，计算效率会相对较低。通过基于对几何形状的合理描述，设计对应的支撑映射函数，能够有效的解决检测精度和计算效率的问题，例如，针对一些基本的几何形状，例如球体、标准圆柱、立方体等，其支撑映射函数计算较为简单，且所消耗的计算资源较小，但是当需要拟合的目标拟合部位的形状不是标准几何体时，通过标准几何体进行拟合时的精度相对较差，会导致碰撞检测的误差增大。本申请中，提出通用圆柱的概念，相较于传统的圆柱其具有较多的参数可供调整，能够更好的适应不同形状的机器人，通过合理的设计相应的支撑映射函数，可以兼顾检测精度和计算效率。
具体的，步骤S3根据目标参数计算通用圆柱的支撑映射函数包括：
S31、确定通用圆柱的局部坐标系，将机器人基坐标系下的目标方向v基于预设转换关系转换为局部坐标系下的转换方向w；
S32、基于局部坐标系确定转换方向w的投影最远点M；
S33、基于预设转换关系将投影最远点M转换为基于机器人基坐标系的目标坐标；
其中，参图2a，图2a中的通用圆柱示例地展示了局部坐标系的构建，支撑映射函数将向量映射到物体C上，其满足如下条件：SC(v)∈C,v·SC(v)＝max{v·x:x∈C}，其中v是机器人基坐标系下的目标方向，SC(v)是对应的支撑点，通过分析验证，通用圆柱的支撑点SC(v)处于其上底面和/或下底面，将机器人的目标方向v基于预设的转换方向转换为 局部坐标系的转换方向w，所述预设的转换方向即机器人基坐标系和通用圆柱的局部坐标系之间的转换关系，可由机器人厂商预先设定，将机器人基坐标系下的目标方向转换至通用圆柱的局部坐标系，求取投影最远点M，再将投影最远点转换为基于机器人基坐标系的目标坐标。
针对通用圆柱，其与N棱柱的构型较为接近，当N取无穷时则为通用圆柱，基于相关资料的研究，当给定任一方向时，N棱柱对应的支撑点SC(v)必然在棱柱的顶点上，同理，对于通用圆柱而言，其支撑点必然存在于上底面和/或下表面。
进一步的，步骤S32基于局部坐标系确定转换方向w的投影最远点M包括：
S321、确定转换方向w＝(wx,wy,wz)，则通用圆柱上底面任一点的坐标为(a1cos(θ),b1sin(θ),h/2)，下底面任一点坐标为其中a1、b1、a2、b2、h是通用圆柱参数，wx、wy、wz分别是转换方向w在局部坐标系x、y、z轴方向上的投影，θ和是角度变量；
S322、确定上底面在转换方向w上的投影值为 下表面在转换方向w上的投影值为
S323、确定上底面的最大投影值对应点的解析解为：θmax＝argmaxθ(p1(θ))，下表面的最大投影值对应点的解析解为：
S324、将上底面的最大投影值p1(θmax)和下表面的最大投影值中的较大值确定为投影最远点M。
至此，根据通用圆柱适应性的计算了支撑映射函数，执行步骤S4，根据通用圆柱的支撑映射函数，采用GJK算法进行机器人的碰撞检测。
GJK(Gilbert-johnson-Keerthi)算法是一种进行碰撞检测的算法，能够实现空间中凸的几何形状间的碰撞检测，GJK算法依赖于支撑映射函数的构建，通过判断两个多边形/体之间是否存在两个坐标相减后为原点来判断是否发生碰撞，GJK依赖于支撑映射函数计算的过程计算效率高且消耗资源小。
具体的，步骤S4采用GJK算法进行机器人的碰撞检测前还包括：S41、确定机器人待拟合部位中除目标拟合部位之外的其他待拟合部位，根据标准几何体对其他待拟合部位进行形状拟合，确定其他待拟合部位的形状拟合参数并计算对应的支撑映射函数。机器人完成对所有待拟合部位的形状拟合后，根据对应的支撑映射函数和GJK算法可执行碰撞检测。其中，标准几何体的支撑映射函数的计算现有技术中已有涉及，此处不再展开，将所有的 拟合过程完成后，生成了碰撞检测的拟合形状集合，从而根据GJK算法进行碰撞检测。
具体的，步骤S4采用GJK算法进行机器人的碰撞检测前还包括：根据对待拟合部位的形状拟合生成机器人碰撞检测的拟合形状集合，确定其中两个形状为碰撞检测对象，采用GJK算法在三维空间中检测所述两个形状是否存在重叠、相交或分离，检测存在重叠或相交时确定为发生碰撞。可理解的，对机器人的形状拟合包括可选的对机器人本体、本体所连接的负载以及机器人环境的拟合，根据所有的拟合结果生成形状拟合，每次分别对拟合形状集合中的两个形状进行碰撞检测，以确定两个形状间是否存在碰撞。
以上本方案的优选实施例，定义了通用圆柱的类型，相较于传统的标准几何体，通用圆柱上底面和下底面的形状易于调节，更能适应市面上的机器人构造，尤其是适用于协作机器人的构造，对机器人的准确拟合能够保证碰撞检测的精度。同时，基于GJK算法，设计针对通用圆柱的支撑映射函数，计算效率高且消耗资源小。该实施例所提供的碰撞检测方法，不依赖于传感器的检测结果，通过精确的形状拟合，简化机器人硬件要求，同时能够保证碰撞检测的准确性。
在示例性实施例中，本申请还提供了一种存储有计算机程序的计算机可读存储介质，例如存储有计算机程序的存储器，所述计算机程序可由处理器执行以完成机机器人碰撞检测方法。可选地，存储介质可以是非临时性计算机可读存储介质，例如，所述非临时性计算机可读存储介质可以是ROM、随机存取存储器(RAM)、CD-ROM、磁带、软盘和光数据存储设备等。
在示例性实施例中，本申请还提供了一种电子设备，该电子设备包括存储器和处理器，所述存储器存储有计算机程序；所述处理器用于执行所述存储器中的所述计算机程序，以实现上述的机器人碰撞检测方法的各步骤。
请参见图5，在一些实施例中，电子设备500可以包括处理器510、存储器520、输入/输出部件530和通信端口540。处理器(例如，CPU)510可以以一个或多个处理器的形式执行程序指令。存储器520包括不同形式的程序存储器和数据存储器，例如，硬盘、只读存储器(ROM)、随机存储器(RAM)等，用于存储由计算机处理和/或传输的各种各样的数据文件。输入/输出部件530可以用于支持处理设备与其他部件之间的输入/输出。通信端口540可以与网络连接，用于实现数据通信。示例性的处理设备可以包括存储在只读存储器(ROM)、随机存储器(RAM)和/或其他类型的非暂时性存储介质中的由处理器510执行的程序指令。本说明书实施例的方法和/或流程可以以程序指令的方式实现。
最后还需要指出，由于文字表达的有限性，上述说明仅是示例性的，并非穷尽性的， 本发明并不限于所披露的各实施方式，在不偏离上述示例的范围和精神的情况下，对于本领域的技术人员来说还可以作若干改进和修饰，这些改进和修饰也应视为本发明的保护范围。因此本发明的保护范围应以权利要求为准。

Claims (12)

  1. 一种机器人碰撞检测方法，其特征在于，包括：
    S1、预先定义通用圆柱和通用圆柱参数，所述通用圆柱包括上底面、下底面和柱面，所述上底面平行于下底面，上底面和下底面的中心点连线垂直于上底面和下底面，所述通用圆柱参数包括上底面的长轴半径a1和短轴半径b1、下底面的长轴半径a2和短轴半径b2、以及柱面高度h；
    S2、确定通用圆柱的目标拟合部位，调整通用圆柱参数以对目标拟合部位的形状进行拟合，将完成拟合时的通用圆柱参数确定为目标参数；
    S3、根据所述目标参数计算通用圆柱的支撑映射函数；
    S4、根据通用圆柱的支撑映射函数，采用GJK算法进行机器人的碰撞检测。
  2. 根据权利要求1所述的机器人碰撞检测方法，其特征在于，所述机器人被配置为包括几何形状拟合库，所述几何形状拟合库包括所述通用圆柱和标准几何体，所述标准几何体包括球体、圆柱体、立方体和锥体中的至少部分。
  3. 根据权利要求2所述的机器人碰撞检测方法，其特征在于，步骤S2确定通用圆柱的目标拟合部位包括：
    S21、获取机器人和/或机器人工作环境的结构模型；
    S22、将所述结构模型拆分为多个待拟合部位；
    S23、根据所述待拟合部位确定适合根据通用圆柱进行形状拟合的目标拟合部位。
  4. 根据权利要求3所述的机器人碰撞检测方法，其特征在于，步骤S21获取机器人的结构模型包括：
    获取机器人的本体结构模型和/或获取机器人本体所连接的负载结构模型。
  5. 根据权利要求3所述的机器人碰撞检测方法，其特征在于，步骤S23根据所述待拟合部位确定适合根据通用圆柱进行形状拟合的目标拟合部位包括：
    获取用户的操作指令以确定目标拟合部位，所述操作指令基于待拟合部位选择目标拟合部位；
    和/或，根据预设的形状选择方法确定目标拟合部位，所述预设的形状选择方法被配置为基于待拟合部位选择目标拟合部位。
  6. 根据权利要求5所述的机器人碰撞检测方法，其特征在于，所述根据预设的形状选择方法确定目标拟合部位包括：
    确定待拟合部位与通用圆柱的形状匹配度，根据所述形状匹配度将符合预设匹配条件的待拟合部位确定为目标拟合部位。
  7. 根据权利要求3所述的机器人碰撞检测方法，其特征在于，步骤S4采用GJK算法进行机器人的碰撞检测前还包括：
    S41、确定机器人待拟合部位中除目标拟合部位之外的其他待拟合部位，根据标准几何体对所述其他待拟合部位进行形状拟合，确定其他待拟合部位的形状拟合参数并计算对应的支撑映射函数。
  8. 根据权利要求3所述的机器人碰撞检测方法，其特征在于，步骤S4采用GJK算法进行机器人的碰撞检测前还包括：
    根据对待拟合部位的形状拟合生成机器人碰撞检测的拟合形状集合，确定其中两个形状为碰撞检测对象，采用GJK算法在三维空间中检测所述两个形状是否存在重叠、相交或分离，检测存在重叠或相交时确定为发生碰撞。
  9. 根据权利要求1所述的机器人碰撞检测方法，其特征在于，步骤S3根据所述目标参数计算通用圆柱的支撑映射函数包括：
    S31、确定通用圆柱的局部坐标系，将机器人基坐标系下的目标方向v基于预设转换关系转换为通用圆柱的局部坐标系下的转换方向w；
    S32、基于局部坐标系确定转换方向w的投影最远点M；
    S33、基于预设转换关系将投影最远点M转换为基于机器人基坐标系的目标坐标。
  10. 根据权利要求9所述的机器人碰撞检测方法，其特征在于，步骤S32基于局部坐标系确定转换方向w的投影最远点M包括：
    S321、确定转换方向w＝(wx,wy,wz)，则通用圆柱上底面任一点的坐标为(a1cos(θ),b1sin(θ),h/2)，下底面任一点坐标为其中a1、b1、a2、b2、h是通用圆柱参数，wx、wy、wz分别是转换方向w在局部坐标系x、y、z轴方向上的投影，θ和是角度变量；
    S322、确定上底面在转换方向w上的投影值为 下表面在转换方向w上的投影值为
    S323、确定上底面的最大投影值对应点的解析解为：θmax＝argmaxθ(p1(θ))，下表面的最大投影值对应点的解析解为：
    S324、将上底面的最大投影值p1(θmax)和下表面的最大投影值中的较大值确定为投影最远点M。
  11. 一种计算机可读存储介质，存储有计算机程序，其特征在于，所述计算机程序被处 理器执行时实现权利要求1至10中任意一项所述的机器人碰撞检测方法。
  12. 一种电子设备，其特征在于，包括：
    存储器，所述存储器存储有计算机程序；
    处理器，用于执行所述存储器中的所述计算机程序，以实现权利要求1至10中任一项所述机器人碰撞检测方法。
PCT/CN2023/092493 2022-05-09 2023-05-06 一种机器人碰撞检测方法、存储介质及电子设备 WO2023217032A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210496159.1 2022-05-09
CN202210496159.1A CN114872043B (zh) 2022-05-09 2022-05-09 一种机器人碰撞检测方法、存储介质及电子设备

Publications (1)

Publication Number Publication Date
WO2023217032A1 true WO2023217032A1 (zh) 2023-11-16

Family

ID=82673233

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/092493 WO2023217032A1 (zh) 2022-05-09 2023-05-06 一种机器人碰撞检测方法、存储介质及电子设备

Country Status (2)

Country Link
CN (1) CN114872043B (zh)
WO (1) WO2023217032A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114872043B (zh) * 2022-05-09 2023-11-17 苏州艾利特机器人有限公司 一种机器人碰撞检测方法、存储介质及电子设备

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105137973A (zh) * 2015-08-21 2015-12-09 华南理工大学 一种人机协作场景下的机器人智能躲避人类方法
US9671777B1 (en) * 2016-06-21 2017-06-06 TruPhysics GmbH Training robots to execute actions in physics-based virtual environment
CN109500811A (zh) * 2018-11-13 2019-03-22 华南理工大学 一种面向人机共融的机器人主动避让人类的方法
CN111360824A (zh) * 2020-02-27 2020-07-03 中科新松有限公司 一种双臂自碰撞检测方法和计算机可读存储介质
CN113244622A (zh) * 2021-06-22 2021-08-13 腾讯科技（深圳）有限公司 碰撞数据处理方法和装置、存储介质及电子设备
CN114872043A (zh) * 2022-05-09 2022-08-09 苏州艾利特机器人有限公司 一种机器人碰撞检测方法、存储介质及电子设备

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103722565B (zh) * 2014-01-23 2015-09-16 哈尔滨工业大学 仿人机器人自碰撞监控系统及监控方法
CN204162934U (zh) * 2014-09-26 2015-02-18 广州东奥电气有限公司 一种多功能快速干衣柜
CN105512377B (zh) * 2015-11-30 2017-12-12 腾讯科技（深圳）有限公司 实时虚拟场景中圆柱碰撞体与凸体碰撞探测的方法和系统
CN107116540B (zh) * 2016-02-24 2019-07-02 中国科学院沈阳计算技术研究所有限公司 一种基于scs包围结构的机器人碰撞检测方法
DE102019103349B3 (de) * 2019-02-11 2020-06-18 Beckhoff Automation Gmbh Industrierobotersystem und Verfahren zur Steuerung eines Industrieroboters
CN111203916B (zh) * 2020-01-16 2021-06-08 山西万合智能科技有限公司 一种机器人关节与几何体碰撞的检测方法
CN111504328B (zh) * 2020-05-22 2022-07-08 梅卡曼德（北京）机器人科技有限公司 机器人运动规划方法、路径规划方法、抓取方法及其装置
CN113664832A (zh) * 2021-09-01 2021-11-19 上海节卡机器人科技有限公司 一种机器人碰撞预测方法、计算机存储介质及电子设备

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105137973A (zh) * 2015-08-21 2015-12-09 华南理工大学 一种人机协作场景下的机器人智能躲避人类方法
US9671777B1 (en) * 2016-06-21 2017-06-06 TruPhysics GmbH Training robots to execute actions in physics-based virtual environment
CN109500811A (zh) * 2018-11-13 2019-03-22 华南理工大学 一种面向人机共融的机器人主动避让人类的方法
CN111360824A (zh) * 2020-02-27 2020-07-03 中科新松有限公司 一种双臂自碰撞检测方法和计算机可读存储介质
CN113244622A (zh) * 2021-06-22 2021-08-13 腾讯科技（深圳）有限公司 碰撞数据处理方法和装置、存储介质及电子设备
CN114872043A (zh) * 2022-05-09 2022-08-09 苏州艾利特机器人有限公司 一种机器人碰撞检测方法、存储介质及电子设备

Also Published As

Publication number Publication date
CN114872043A (zh) 2022-08-09
CN114872043B (zh) 2023-11-17

Similar Documents

Publication Publication Date Title
CN108705532B (zh) 一种机械臂避障路径规划方法、设备及存储设备
Liu et al. Grasping posture of humanoid manipulator based on target shape analysis and force closure
WO2020206666A1 (zh) 基于散斑图像的深度估计方法及装置、人脸识别系统
WO2023217032A1 (zh) 一种机器人碰撞检测方法、存储介质及电子设备
US9747721B2 (en) Technique for filling holes in a three-dimensional model
WO2022000260A1 (zh) 地图的更新方法、装置、可移动平台及存储介质
TWI748409B (zh) 資料處理方法、處理器、電子設備和電腦可讀媒體
WO2020093950A1 (zh) 三维物体分割方法、设备和介质
EP4102458A1 (en) Method and apparatus for identifying scene contour, and computer-readable medium and electronic device
WO2022021156A1 (zh) 用于机器人抓取三维物体的方法和装置
WO2022217794A1 (zh) 一种动态环境移动机器人的定位方法
CN116152444B (zh) 基于数字孪生的三维场景模型自动吸附方法、装置及介质
CN104240290A (zh) 一种基于包围球模型的三维电缆相交检测方法
WO2022148349A1 (zh) 人脸追踪方法、装置、电子设备和存储介质
WO2022028110A1 (zh) 自移动设备的地图创建方法、装置、设备及存储介质
WO2023231640A1 (zh) 动态沿边路径生成方法、装置、计算机设备和存储介质
KR102607015B1 (ko) 실시간 3d 센서 데이터 분석을 통한 용접선 자동인식 방법, 장치 및 컴퓨터프로그램
WO2023005195A1 (zh) 地图数据的处理方法、装置、家用电器和可读存储介质
CN116342704A (zh) 一种扫描点类别确定方法、装置、计算机设备及存储介质
TWI406189B (zh) 點雲三角網格面構建方法
CN115861547A (zh) 一种基于投影的模型表面样条线生成方法
Shiang et al. Estimate of minimum distance between convex polyhedra based on enclosed ellipsoids
JP2019030930A (ja) 相対的に移動する物体間の最近接点計算方法及び干渉評価方法
CN113768535A (zh) 一种遥操作用超声仿型探头姿态自校准方法、系统及装置
Ju et al. Fast and accurate collision detection based on enclosed ellipsoid

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23802804

Country of ref document: EP

Kind code of ref document: A1